productivity, the net physical value of a child at birth, discounted by Enke's 10 to 15 percent per year, still yields a substantial negative current value.* Enke concludes that this "social gain" from preventing a birth sets a ceiling on the real resources a society can spend to avert births.
Though Enke's endorsement of various schemes to exhort and to bribe parents into avoiding additional births may be sound in countries with abundant unskilled labor and little human and physical capital, his analytical method for supporting these policy conclusions is inconsistent with the central tenet of welfare economics-that people are the best judges of their own welfare. Enke's analysis neglects altogether the nonpecuniary returns to children (44). Clearly, children represent much more to parents than producer goods.
Viewed only as producer goods, children do appear to be a poor social investment. But the rest of the society is not usually asked to bear most of the costs of a child's consumption; parents are willing and able, in most cases, to bear these pecuniary burdens of child dependency in return for intangible parental rewards. For the society to justify policies that seek to change the preferred reproductive patterns of parents, children must also represent a burden that transcends the family unit and poses costs to others than their own kin. On this score, Enke has marshalled no evidence whatsoever that children are a source of external, social diseconomies that warrant social expenditures and interpersonal transfers to reduce birth rates. Enke's computation that a child has a negative net physical value at birth is therefore no indication that society and its constituent members would be better off with fewer births.
The second approach to evaluating the gains of slowing population growth has received its most elaborate treatment by Coale and Hoover in their pioneering study of India (45). A reduction in the birth rate reduces the proportion of the population that is not yet of the age when it can participate in the labor force and contribute directly to production. If aggregate income is unaffected by the reduced birth rate (at least for 10 or 15 years), per capita income must rise in comparison with the case of constant birth rates. If savings rates rise in response to this relative increase in per capita income, further long-term advantages accrue to the population experiencing the declining birth rate. This methodology presumes that maximizing per capita income (or consumption) is the appropriate social objective and that simple aggregate production functions adequately describe long-term growth pros-
*Ohlin (43) states this conclusion which was confirmed by the author. In a later paper by Ohlin, summarized by Easterlinc (25), it appears that Ohlin has shown that under reasonable assumptions concerning the character of the aggregate production function, children are not a good investment unless the marginal return on capital falls to the rate of growth of the population-an implausible situation in today's world of developing nations. Sec also Leibcnstein's critique of Enke's methodology in (44).